I am trying to stream IP Camera with mjpeg codec …

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

**• Hardware Platform (Jetson / GPU)**jetson nano
• DeepStream VersionDS 5.1
• JetPack Version (valid for Jetson only) 4.5.1
• TensorRT Version7.1.13
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)

i tried everything i can get in deepstream . my application actually it’s about streaming 4 IP CAMERA in live stream… i tried so to change the C file but with any success… i still have some trouble with gstreamer plugins
the deepstream-image-data-test application can just deal with local files and the deepstream-test3 application can just deal h.264 codec
and even i tried to with source8_1080_dec for nano config file i didn’t get anything just segmentation fault (core dumped) after some changes relevant to my ip cam … please any clue can help! Environment

3

Hi,
You may try to work out a gst-launch-1.0 command like

$ gst-launch-1.0 rtspsrc ! rtpjpegdepay ! jpegparse ! nvv4l2decoder mjpeg=1 ! nvoverlaysink

And then apply it to deepstream-image-data-test.

4

Hi, thank you for the reply, the command didn’t work but I got this one work
gst-launch-1.0 rtspsrc location=rtsp://*****/mjpeg/1 ! rtpjpegdepay ! jpegparse ! jpegdec ! nvvidconv ! autovideosink
how to apply it on deepstream-image-data-test .?

5

Hi,
Please confirm this command works:

gst-launch-1.0 rtspsrc location=rtsp://*****/mjpeg/1 ! rtpjpegdepay ! jpegparse ! jpegdec ! nvvideoconvert ! 'video/x-raw(memory:NVMM),format=NV12' ! fakesink

If above command works, you can modify the sample to run

rtspsrc location=rtsp://*****/mjpeg/1 ! rtpjpegdepay ! jpegparse ! jpegdec ! nvvideoconvert ! 'video/x-raw(memory:NVMM),format=NV12' ! nvstreammux ! ...
6

Hi, unfortunately, it didn’t work.
I really need help for this task and thank you for your updates.

7

Hi,
Please try

gst-launch-1.0 rtspsrc location=rtsp://*****/mjpeg/1 ! rtpjpegdepay ! jpegparse ! jpegdec ! videoconvert ! video/x-raw,format=NV12 ! fakesink

gst-launch-1.0 rtspsrc location=rtsp://*****/mjpeg/1 ! rtpjpegdepay ! jpegparse ! jpegdec ! videoconvert ! video/x-raw,format=NV12 ! nvvideoconvert ! 'video/x-raw(memory:NVMM),format=NV12' ! fakesink
8

Hi, I tried them but they both didn’t work than I changed “fakesink” with “autovideosink” in both. I got with first command pipeline black and white screen… and with the second command worked but with bad resolution… it seems like it can not accept “fakesink”

9

Hi,
You may break the pipeline and try:

gst-launch-1.0 rtspsrc location=rtsp://*****/mjpeg/1 ! rtpjpegdepay ! jpegparse ! jpegdec ! fakesink

gst-launch-1.0 rtspsrc location=rtsp://*****/mjpeg/1 ! rtpjpegdepay ! jpegparse ! jpegdec ! videoconvert ! video/x-raw,format=NV12 ! fakesink

Please try to work out a working gst-launch-1.0 command. And then customize existing samples based on the command.

10

Hi, thank you for the reply again but I am sorry to tell you that these commands didn’t work again… I’m actually confused with these pipelines … will be able to use the only gst-launch-1.0 working command without “fakesink” and how to customize it

gst-launch-1.0 rtspsrc location=rtsp://*****/mjpeg/1 ! rtpjpegdepay ! jpegparse ! jpegdec ! nvvidconv ! autovideosink